Post on 17-Mar-2020
transcript
MIC2168A
1MHz PWM Synchronous Buck-Control IC
General Description The MIC2168A is a high-efficiency, simple to use 1MHz PWM synchronous buck-control IC housed MSOP-10 package. The MIC2168A allowDC/DC solutions with a minimal external count and cost.
in a small s compact component
Micrel Inc. • 2180 Fortune Drive • San Jose, CA 95131 • USA • tel +1 (408) 944-0800 • fax + 1 (408) 474-1000 • http://www.micrel.com
January 2010 M9999-011510
e
nnel MOSFET, eliminating th
e
internal in-rush-current limiting
identical to the MIexception that the MIC2168A increaseblanking time from 80ns (typ.) to 120n
with a wide junction operating range of –40°C to +125°C. All support documentation can be found on Micrel’s web site at www.micrel.com
The MIC2168A operates from a 3V to 1without the need of any additional bias voloutput voltage can be precisely regulated doThe adaptive all N-Channel MOSFET drivallows efficiencies over 95% across a wide loaThe MIC2168A senses current across the hCha
4.5V input, tage. Th
wn to 0.8V. scheme e
d range. igh-side N-
e need for an urrent limit expensive and lossy current-sense resistor. C
accuracy is maintained by a positive t mperature coefficient that tracks the increasing RDSexternal MOSFET. Further cost and space by the
(ON) of the are saved
and digital soft-start. The MIC2168A is C2168 with the
s the overcurrent s (typ.)
The MIC2168A is available in a 10-pin MSOP package,
Features
.
14.5V input voltage range n to 0.8V
ciency ion
s high-side N-Channel
sistor ses efficiency
se with hysteretic transient recovery ode
ction protects the load in fault conditions
rrent limit speeds up recovery time Hiccup mode short-circuit protection
rnal soft-start ize MSOP 10-pin package
ns on
• Set-top boxes • Graphic cards • LCD power supplies • Telecom power supplies • Networking power supplies • Cable modems and routers
_______________________________________________________________________________________________________
Typical Application
• 3V to• Adjustable output voltage dow• Up to 95% effi• 1MHz PWM operat• Adjustable current-limit sense
MOSFET current • No external current sense re• Adaptive gate drive increa• Ultra-fast respon
m• Overvoltage prote
• Dual mode cu• • Inte• Small-s
Applicatio• Point-of-load DC/DC conversi
Micrel MIC2168A
January 2010 2 M9999-011510
Ordering Information Part Number Standard Pb-Free
Frequency Junction Temperature Range(1) Package
MIC2168ABMM MIC2168AYMM 1MHz –40° to +125°C 10-Pin MSOP
Pin Configuration
FB 5 GND6
1VIN
VDD
CS
COMP/EN
10 BST
HSD
VSW
LSD
9
8
7
2
3
4
MIC2168A 10-Pin MSOP (MM)
Pin Description
Pin Number Pin Name Pin Function 1 VIN Supply Voltage (Input): 3V to 14.5V. 2 VDD 5V Internal Linear Regulator (Output): VDD is the external MOSFET gate drive supply voltage
and an internal supply bus for the IC. When VIN is <5V, this regulator operates in dropout mode.
3 CS Current Sense / Enable (Input): Current-limit comparator noninverting input. The current limit is sensed across the MOSFET during the ON time. The current can be set by the resistor in series with the CS pin.
4 COMP Compensation (Input): Dual function pin. Pin for external compensation. If this pin is pulled below 0.2V, with the reference fully up, the device shuts down (50μA typical current draw).
5 FB Feedback (Input): Input to error amplifier. Regulates error amplifier to 0.8V. 6 GND Ground (Return). 7 LSD Low-Side Drive (Output): High-current driver output for external synchronous MOSFET. 8 VSW Switch (Return): High-side MOSFET driver return. 9 HSD High-Side Drive (Output): High-current output-driver for the high-side MOSFET. When VIN is
between 3.0V to 5V, 2.5V threshold-rated MOSFETs should be used. At VIN > 5V, 5V threshold MOSFETs should be used.
10 BST Boost (Input): Provides the drive voltage for the high-side MOSFET driver. The gate-drive voltage is higher than the source voltage by VIN minus a diode drop.
Micrel MIC2168A
January 2010 3 M9999-011510
e M atingSupply Voltage (VIN) ........................................... 15.5V Bootstrapped Voltage (VBST) ............................VIN +5V
re (TJ).......... –40°C ≤ TJ ≤ +125°C (TS) ................ –65°C to +150°C
rating RSupply Voltage (VIN) ............................. +3V to +14.5V Output Voltage Range ....................0.8V to VIN × DMAX Package Thermal Resistance 10-Pin MSOP (θJA) .................................. 180°C/W
Electrical Characteristics(3) TJ = 25°C, VIN = 5V, unless otherwise specified. Bold values indicate –40°C < TJ < +125°C
Parameter Condition Min. Typ. Max. Units
Absolut aximum R s(1)
Junction TemperatuStorage Temperature
Ope atings(2)
Feedback Voltage Reference (± 1%) 0.792 0.8 0.808 V Feedback Voltage Reference (± 2% over temp) 0.784 0.8 0.816 V Feedback Bias Current 30 100 nA Output Voltage Line Regulation 0.03 %/V Output Voltage Load Regulation 0.5 % Output Voltage Total Regulation 3V ≤ VIN ≤ 14.5V; 1A ≤ IOUT ≤ 10A; (VOUT = 2.5V)(4) 0.6 % Oscillator Section Oscillator Frequ 900 1000 1100 kHz ency Maximum Duty Cycle 90 Minimum On-Time(4) 30 60 ns Input and VDD Supply PWM Mode Supply Curre 1.6 3 mA nt VCS = VIN –0.25V; VFB = 0.7V
(output switching but excluding external MOSFET gate current.)
Shutdown Quiescent Curr 50 150 µA ent VCOMP/EN=0V VCOMP Shutd hres 0.1 0.25 0.4 V own T hold VCOMP Shutdo lankiPeriod
4 ms wn B ng CCOMP = 100nF
Digital Supply Voltage (V 5 5.3 V DD) VIN ≥ 6V 4.7 Error Amplifier DC Gain 70 dB Transconductance 1 ms Soft Start Soft-Start Current After timeout of internal timer.
(See “Soft-Start” section.) 8.5 µA
Current Sense CS Overcurrent Trip Point VCS = VIN –0.25V 160 200 240 µA Temperature Coefficient +1800 ppm/°C
Notes: 1. Absolute maximum ratings indicate limits beyond which damage to the component may occur. Electrical specifications do not apply when
operating the device outside of its operating ratings. The maximum allowable power dissipation is a function of the maximum junction temperature, TJ(max), the junction-to-ambient thermal resistance, θJA, and the ambient temperature, TA. The maximum allowable power dissipation will result in excessive die temperature, and the regulator will go into thermal shutdown.
2. Devices are ESD sensitive, handling precautions required. 3. Specification for packaged product only. 4. Guaranteed by design.
Micrel MIC2168A
January 2010 4 M9999-011510
(5) ntinified. B Typ. Max. Units
Electrical Characteristics (co ued) T = 25°C, V = 5V, unless otherwise specJ IN old values indicate –40°C < TJ < +125°C
Parameter Condition Min. Output Fault Correction Thresholds Upper Threshold, VFB_OVT (relative to VFB) +3 % Lower Threshold, VFB_UVT (relative to VFB) -3 % Gate Drivers Rise/Fall Time Into 3000pF at VIN > 5V 30 ns
Source, VIN = 5V 6 Ω Sink, VIN = 5V 6 Ω Source, VIN = 3V 10 Ω
Output Driver Impedance
IN 10 Ω Sink, V = 3V Driver Non-Overlap Time Note 6 10 20 ns
Notes: 5. Specification for packaged product only.
. 6. Guaranteed by design
Micrel MIC2168A
Typical Characteristics
January 2010 5 M9999-011510
Micrel MIC2168A
January 2010 6 M9999-011510
Typical Characteristics (continued)
Micrel MIC2168A
January 2010 7 M9999-011510
Functional Diagram
MIC2168A Block Diagram
Micrel MIC2168A
January 2010 8 M9999-011510
Functional Description The MIC2168A is a voltage mode, synchronous step-down switching regulator controller designed for high output power without the use of an external sense resistor. It includes an internal soft-start function which reduces the power supply input surge current at start-up by controlling the output voltage rise time, a PWM generator, a reference voltage, two MOSFET drivers, and short-circuit current limiting circuitry to form a complete 1MHz switching regulator.
Theory of Operation The MIC2168A is a voltage mode step-down regulator. The block diagram illustrates the voltage control loop. The output voltage variation due to load or line changes will be sensed by the inverting input of the transconductance error amplifier via the feedback resistors R3, and R2 and compared to a reference voltage at the non-inverting input. This will cause a small change in the DC voltage level at the output of the error amplifier which is the input to the PWM comparator. The other input to the comparator is a 0 to 1V triangular waveform. The comparator generates a rectangular waveform whose width tON is equal to the time from the start of the clock cycle t0 until t1, the time the triangle crosses the output waveform of the error amplifier. To illustrate the control loop, let us assume the output voltage drops due to sudden load turn-on, this cause the inverting input of the errdivided down version of VOUT to be slightly less than the reference voltage causing the output voltage of the error amplifier to go high. This will cause the PWM comparator to increase tON time of the top side MOSFET, causing the output voltage to go up and bringing VOUT back in regulation.
Soft-Start The COMP pin on the MIC2168A is used for the following three functions:
1. Disables the part by grounding this pin 2. External compensation to stabilize the voltage
control loop 3. Soft-start
For better understanding of the soft-start feature, let’s assume VIN = 12V, and the MIC2168A is allowed to power-up. The COMP pin has an internal 8.5μA current source that charges the external compensation capacitor. As soon as this voltage rises to 180mV (t = Cap_COMP × 0.18V/8.5μA), the MIC2168A allows the internal VDD linear regulator to power up and as soon as it crosses the undervoltage lockout of 2.6V, the chip’s internal oscillator starts switching. At this point in time,
the COMP pin current source increases to 40μA and an internal 11-bit counter starts counting which takes approximately 2ms to complete. During counting, the COMP voltage is clamped at 0.65V. After this counting cycle the COMP current source is reduced to 8.5A and the COMP pin voltage rises from 0.65V to 0.95V, the bottom edge of the saw-tooth oscillator. This is the beginning of 0% duty cycle and it increases slowly causing the output voltage to rise slowly. The MIC2168A has two hysteretic comparators that are enabled when VOUT is within ±3% of steady state. When the output voltage reaches 97% of programmed output voltage, then the gm error amplifier is enabled along with the hysteretic comparator. This point onwards, the voltage control loop (gm error amplifier) is fully in control and will regulate the output voltage. Soft-start time can be calculated approximately by adding the following four time frames:
t1 = Cap_COMP × 0.18V/8.5μA t2 = 12 bit counter, approx 2ms t3 = Cap_COMP × 0.3V/8.5μA
would or amplifier which is
8.5μ.Cap_COMP0.5
VV
t4IN
OUT ××⎟⎟⎠
⎞⎜⎜⎝
⎛
Soft-Start Time(Cap_COMP = 100nF) = t1 + t2 + t3 + t4 = 2.1ms + 2ms + 3.5ms + 1.8ms = 10ms
Current Limit The MIC2168A uses the RDS(ON) of the top power MOSFET to measure output current. Since it uses the drain to source resistance of the power MOSFET, it is not very accurate. This scheme is adequate to protect the power supply and external components during a fault condition by cutting back the time the top MOSFET is on if the feedback voltage is greater than 0.67V. In case of a hard short when feedback voltage is less than 0.67V, the MIC2168A discharges the COMP capacitor to 0.65V, resets the digital counter and automatically shuts off the top gate drive, and the gm error amplifier and the –3% hysteretic comparators are completely disabled and the soft-start cycles restarts. This mode of operation is called the “hiccup mode” and its purpose is to protect the down stream load in case of a hard short. The circuit in Figure 1 illus tes the MIC2168A current limiting circuit.
tra
Micrel MIC2168A
Figure 1. The MIC2168A Current Limiting Circ
The current limiting resistor RCS is calculafollowing equation:
January 2010 9 M9999-011510
u
ted by the
it
200µAIR
R LDS(ON)Q1CS
×=
( )Current Ripple Inductor21II LOADL ==
where:
Inductor Ripple Current = ( )
LFV SWITCHINGIN ××VV
V OUTINOUT
−=
FSWITCHING = 1MHz
o
0% with add a 50% e equation d MOSFET mended to
connect RCS resistor directly to the drain of the top MOSFET Q1, and the RSW resistor to the source of Q1 to accurately sense the MOSFETs RDS(ON). To make the MIC2168A insensitive to board layout and noise, a 1.4Ω resistor and a 1000pF capacitor is recommended below the switch node and ground. A 0.1μF capacitor in parallel with RCS should be connected to filter some of the switching noise.
ally generates VDD for to the gate drives. This low-dropout regulator
y greater than 5V. For VDD linear regulator is out. Therefore, it is supply to the input input supplies between
igh-side drive circuit is designed to annel MOSFET. The block diagram on bootstrap circuit, consisting of D1 and
rgy to the high-side drive circuit. Capacitor CBST is charged while the low-side MOSFET is on and the voltage on the VSW pin is approximately
ver is turned on, turn the MOSFET on. As
voltage on the VSW pin increases to approximately VIN. Diode D1 is reversed
ile continuing to keep the low-side switch is rged through D1. The internal 5V VDD bias te drive voltage is 5V ate drive voltage is ge drop across D1. An the high- and low-side prevent current from eded through both
The MIC2168A controller works from input voltages of 3V to 13.2V and has an internal 5V regulator to provide power to turn the external N-Channel power MOSFETs for high- and low-side switches. For applications where VIN < 5V, the internal VDD regulator operates in dropout mode, and it is necessary that the power MOSFETs used are low threshold and are in full conduction mode for VGS of 2.5V. For applications when VIN > 5V; logic-level MOSFETs, whose operation is specified at VGS = 4.5V must be used. It is important to note the on-resistance of a MOSFET increases with increasing temperature. A 75°C rise in junction temperature will increase the channel resistance of the MOSFET by 50% to 75% of the resistance specified at 25°C. This change in resistance must be accounted for when calculating MOSFET power dissipation and in calculating the value of current-sense (CS) resistor.
200μA is the internal sink current to prMIC2168A current limit. The MOSFET RDS(ON) varies 30% to 4temperature; therefore, it is recommended tomargin to the load current (ILOAD) in the abovto avoid false current limiting due to increasejunction temperature rise. It is also recom
Internal VDD Supply The MIC2168A controller internself biasing and to provide power VDD supply is generated through aand generates 5V from VIN supplsupply voltage less than 5V, theapproximately 200mV in droprecommended to short the VDDsupply through a 10Ω resistor for 2.9V to 5V.
gram the
biased and CBST floats high whthe high-side MOSFET on. Whenturned back on, CBST is rechadrive voltage is derived from thesupply. The nominal low-side gaand the nominal high-side gapproximately 4.5V due the voltaapproximate 20ns delay betweendriver transitions is used to simultaneously flowing unimpMOSFETs.
MOSFET Selection
MOSFET Gate Drive The MIC2168A hswitch an N-ChPage 6 shows aCBST, supplies ene
0V. When the high-side MOSFET drienergy from CBST is used tothe MOSFET turns on, the
Micrel MIC2168A
January 2010 10 M9999-011510
Total gate charge is the charge required toMOSFET on and off under specified conditions (VDS and VGS). The gate charge isby the MIC2168A gate-drive circuit. At 1MHzfrequency and above, the gate charge csignificant source of power dissipation in the MAt low output load, this power dissipation is as a reduction in efficiency. The average current required to drive the MOSFET is:
where:
IG[high-side](avg) = average high-side MOSFEcurrent.
total gate charge for the high-side MOSF
from manufacturer’s data sheet for VGS = 5V. T MOSFET is turned on and off abecause the freewheelin diode is conductithis time. The switching loss for the low-side Mu Also, the gate-drive current for the low-side MOSFET is more accurately calcula
s M
turn the operating supplied
switching an be a IC2168A.
noticeable
high-side
SGside](avg)G[high fQI ×=−
T gate
ET taken
t VDS = 0 ng during OSFET is
QG =
he low-sideg
sually negligible. ted using
CISS at VDS = 0 instead of gate charge. e MFor the low-sid OSFET:
SGSISSside](avg)G[low fVCI ××=−
Since the current from the gate drive comeinput voltage, the power dissipated in thedue to gate drive is:
from the IC2168A
( )side](avg)
SFETs is
G[lowside](avg)G[highINE IIV −− +=GATEDRIVP
A convenient figure of merit for switching MOthe on resistance times the total gate charge RDS(ON) × QG. Lower numbers translate into higher efficiency. Low gate-charge logic-level MOSFETs are a good choice for use with the MIC2168A. Parameters that are important to MOSFET switch selection are:
• Voltage rating • On-resistance • Total gate charge
d bottom MOSFET are tage. A safety factor of
to circuit parasitics. hing transistor is the
during the on-time sses that occur during
the period of time when the MOSFETs turn on and off
The voltage ratings for the top anessentially equal to the input vol20% should be added to the VDS(max) of the MOSFETs to account for voltage spikes due The power dissipated in the switcsum of the conduction losses(PCONDUCTION) and the switching lo
(PAC). PSW = PCONDUCTION + PAC
where:
⎟⎟⎠
⎞⎜⎜⎝
⎛=
−=
+=
×=
IN
O
SW
AC(on)AC(off)AC
2SW(rms)CONDUCTION
VVcycleduty D
the of resistanceonR
PPP
RIP
Making the assumption the transition times are equal; the trapproximated by:
SW
switch MOSFET
turn-on and turn-off ansition times can be
G
OSSGSISST I
CVCt ×+×= INV
where:
ed at VDS = 0 for the MIC2168A)
ss is:
S
ns to 50ns) , typically 0.5V
fS it the switching frequency, nominally 1MHz The low-side MOSFET sw ing losses are negligible and can be ignored for these calculations.
CISS and COSS are measurIG = gate-drive current (1A
The total high-side MOSFET switching lo
PAC = (VIN + VD) × IPK × tT × f where:
tT = switching transition time (typically 20VD = freewheeling diode drop
itch
Micrel MIC2168A
January 2010 11 M9999-011510
r teat
eak-to-peak ripple currents will increapower dissipation in th ductor and MOSFET er output ripple currents will also require more output capacitance to smooth out the larger ripple current.
ller peak-to-peak ripple currents require a larger inductance value and therefore a larger and more e between size, l cost is to set the inductor ripple current to be equal to 20% of the maximum output cur
Inductor Selection Values for inductance, peak, and RMS curequired to select the output inductor. Theoutput voltages and the inductance value depeak-to-peak inducto
rents are input and rmine the lly, higher voltages.
r ripple current. Generinductance values are used with higher inpuLarger p se the
e in s. Larg
Sma
xpensive inductor. A good compromise oss and
rent. The inductance value is calculated by:
( )(max)I0.2f(max)V
V(max)VVLOUTSIN
OUTINOUT×××−×
=
where:
fS = switching frequency, 1MHz 0.2 = ratio of AC ripple current to DC output
The peak-to-peak inductor current (AC ripple curre
current VIN(max) = maximum input voltage
nt) is:
( )Lf(max)VV(max)VVI
SIN
OUTINOUTPP ××
−×=
The peak inductor current is equal to the avercurre
age output e peak-to-peak inductor ripple
IPK = IOUT(max) + 0.5 × IPP The RMS inductor current is used to calculate the I2 × R losses in the inductor.
nt plus one half of thcurrent.
2
(max)IPI1(max)I(rms)INDUCTORI
OUTOUT 3
1⎟⎟⎠
⎞⎜⎜⎝
⎛+×=
e proper selection of he winding resistance. he MIC2168A requires
ll but the most cost st iron powder cores core loss will reduce
the efficiency of the power supply. This is especially at low output power. The winding resistance
decreases efficiency at the higher output current levels. The winding resistance must be minimized although this usually comes at the expense of a larger inductor. The power dissipated in the in r is equal to the sum of the core and copper losses. At higher output loads, the core losses are usually insignificant and can be ignored. At lower output currents, the core losses can be a
information is usually or. Copper loss in the
inductor is calculated by the equation below:
PINDUCTOR Cu = IINDUCTOR(rms)2 × RWINDING
wire, RWINDING, increases f the winding resistance
erating temperature.
where:
e of the wire under operating
T = ambient temperature
RWINDING(20°C) is room temperature winding resistance (usually specified by the manufacturer)
Output Capacitor Selection The output capacitor values are usually determined capacitors ESR (equivalent series resistance). Voltage and RMS current capability are two other important factors selecting the output capacitor. Recommended capacitors are tantalum, low-ESR aluminum electrolytics, and POSCAPS. The output capacitor’s ESR is usually the main cause of output ripple. The output capacitor ESR also affects the overall voltage feedback loop from stability point of view. See “Feedback Loop Compensation” section for more information.
Maximizing efficiency requires thcore material and minimizing tThe high frequency operation of tthe use of ferrite materials for asensitive applications. Lower comay be used but the increase in
noticeable
ducto
significant contributor. Core loss available from the magnetics vend
The resistance of the copper with temperature. The value oused should be at the op
( ))(0042.01 20)20( CHOTCWINDING TT °° −×+×)(hotWINDING RR =
THOT = temperaturload
20°C
Micrel MIC2168A
The maximum value of ESR is calculated:
PP
OUTESR I
ΔVR ≤
where:
VOUT = peak-to-peak output voltage ripIPP = peak-to-peak inductor ripple cur
The total output ripple is
January 2010 12 M9999-011510
ple nt
the ESR d below:
re
a combination of output capacitance. The total ripple is calculate
( )2 ESRPP
2
SOUT
PPOUT RI
fCD)(1IΔV ×+⎟⎟
⎠
⎞⎜⎜⎝
⎛
×−×
=
whe
D = duty cycle
fS = switching frequency The voltage rating of capacitor should be twice the
n c
lated below:
re:
COUT = output capacitance value
voltage for a tantalum and 20% greater for aelectrolytic. The output capacitor RMS calcu
aluminum urrent is
12OUT(rms)
II PPC =
The power dissipated in the output capacitor is
Input Capacitor Selection The input capacitor should be selected for ripple current rating and voltage rating. Tantalum input capacitors may fail when subjected to high inrush currents, caused by turning the input supply on. Tantalum input capacitor voltage rating should be at least 2 times the maximum input voltage to maximize reliability. Aluminum electrolytic, OS-CON, and multilayer polymer film capacitors can handle the higher inrush currents without voltage derating. The input voltage ripple will primarily depend on the input capacitor’s ESR.
to the peak inductor
:
)(2
)( )( OUTrmsOUTOUT CESRCCDISS RIP ×=
The peak input current is equalcurrent, so:
eak)INDUCTOR(pIN RIΔV × )ESR(CIN=
The input capacitor must be rateripple. The RMS value of inpudetermined at the maximum outthe peak-to-peak inductor ripple c
d for the input current t capacitor current is put current. Assuming urrent is low:
D)(1D(max)I(rms)I OUTCIN×≈ −×
The power dissipated in the input capacitor is:
Voltage Setting Components The MIC2168A requires two resistors to set the output voltage as shown in Figure 2.
2
C)DISS(C ININ(rIP ×= )ESR(CIN
Rms)
Figure 2. Voltage-Divider C
where:
onfiguration
VREF for the MIC2168A is typically 0.8V The output voltage is determined by the equation:
⎟⎠
⎞⎜⎝
⎛ +×=R2R11VV REFO
Micrel MIC2168A
January 2010 13 M9999-011510
A typical value of R1 can be between 3k anR1 is too large, it may allow noise to be introduced into the voltage feedback loop. If R1 is too small, in value, it
decrease the efficiency of the power supply, especially at light loads. Once R1 is selected, calc
d 10k. If
will R2 can be
ulated using:
REFVV
R1REFVR2
−
×=
O
External Schottky Diode An external freewheeling diode is used to p the inductor current flow continuous while both MOSFETs
turned off. This dead time prevents cuflowing unimpeded through both MOSFETtypic he diode conducts twice duswitchin average curren through this st be able to handle the peak current.
is:
E(rms) = VIN The power dissi ated by the Schottky diode is:
where:
cessary for c ntains a
rove ise. If the
d to handle as a
relatively slow reverse recovery time and a relatively high forward voltage drop. The power lost in the diode is proportional to the forward voltage drop of the diode. As the high-side MOSFET starts to turn on, the body diode becomes a short circuit for the reverse recovery period, dissipating additional power. The diode recovery and the circuit inductance will cause ringing during the high-side MOSFET turn-on. An external Schottky diode conducts at a lower forward voltage preventing the body diode in the MOSFET from turning on. The lower forward voltage drop dissipates less power than the body diode. The lack of a reverse recovery mechanism in a Schottky diode causes less ringing and less power loss.
Depending on the circuit components and operating conditions, an external Schottky diode will give a 1/2% to 1% improvement in efficiency.
es with an internal sed for compensating
by placing a capacitor (C1) in series with a resistor (R1) and another capacitor C2 in parallel from the CO P pin to ground. See “Functional Block Diagram.”
The power stage of a voltage mode controller has an inductor, L1, with its winding resistance (DCR)
to the output capacitor, COUT, with its as shown in Figure 3.
kee
are rrent from connected s and is ring each
electrical series resistance (ESR) ally 15ns. T
g cycle. Although the ode mu
t diode is small, the di
The reverse voltage requirement of the diode
VDIOD
p
PDIODE = ID(avg) × VF
V = forward F voltage at the peak diode The external Schottky diode, D1, is not necircuit operation since the low-side MOSFETparasitic body diode. The external diode wilefficiency and decrease high frequency noMOSFET body diode is used, it must be ratethe peak and average current. The body diode h
current
ol imp
Feedback Loop Compensation The MIC2168A controller comtransconductance error amplifier uthe voltage feedback loop
M
Power Stage
Figure 3. The Output LC Filter in a Voltage-Mode
Converter The transfer function G(s), for such a system is:
Buck
⎟⎟⎠
⎞⎜⎜⎝ ××
−CsDCR
G(S)⎛
××++××+
××+
CsESR1CLsCsESR(1
2
Plotting this transfer function with the following assumed values (L = 2μH, DCR = 0.009Ω, COUT = 1000μF, ESR = 0.025Ω) gives lot of insight as to why one needs to compensate the loop by adding resistor and capacitors on the COMP pin. Figures 4 and 5 show the gain curve and phase curve for the above transfer function.
Micrel MIC2168A
Figure 4. The Gain Curve for G(s)
January 2010 14 M9999-011510
) and the
Figure 5. Phase Curve for G(s) It can be seen from the transfer function Ggain curve that the output inductor and capaa two pole system with a break frequency at:
(scitor create
OUTC
CLπ21f××
=
Therefore, fLC = 3.6kHz. By looking at the phase curve, it can be seen that theoutput capacitor ESR (0.050Ω) cancels one
of the two
poles (LCOUT) system by introducing a zero at:
OUTZERO CESRf
×××=
π21
Therefore, FZERO = 6.36kHz.
pensating the voltage mended to use higher ESR output
ce they provide a 90° phase gain in the Figure 6, shows
R value of 0.002Ω.
From the point of view of comloop, it is recomcapacitors sinpower path. For comparison purposes,the same phase curve with an ES
Figure 6. The Phase Cur
ve with ESR = 0.002Ω
It can be seen from Figure 5 that at 50kHz, the phase is s Figure 6 where the number
is –150°. This means that the transconductance error amplifier has to provide a phase boost of about 45° to achieve a closed loop phase margin of 45° at a crossover frequency of 50kHz for Figure 4, versus 105°
and C2 compensation fier phase boost
asier to stabilize the using high ESR value
gm Error Amplifier It is undesirable to have high error amplifier gain at high frequencies because high-frequency noise spikes would be picked up and transmitted at large amplitude to the output, thus, gain should be permitted to fall off at high frequencies. At low frequency, it is desired to have high open-loop gain to attenuate the power line ripple. Thus, the error amplifier gain should be allowed to increase rapidly at low frequencies.
approximately –90° versu
for Figure 6. The simple RC scheme allows a maximum error ampliof about 90°. Therefore, it is eMIC2168A voltage control loop byoutput capacitors.
Micrel MIC2168A
January 2010 15 M9999-011510
The transfer function with R1, C1, and Cinternal gm error amplifier can be approximatfollowing equation:
2 for the ed by the
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
⎟⎠⎞
⎜⎝⎛
+
×××++×
××+×=
C2C1
SC2C1R11C2)(C1s
C1SR11mg(s) AmplifierError
The above equation can be simplified by assuming C2<<C1,
⎟⎠⎞
××
×
S)C2
C1
e that R1 ole at the
⎜⎝⎛
+×
×+×=
R1(C1)(1s
SR11mg(s) AmplifierError
From the above transfer function, one can seand C1 introduce a zero and R1 and C2 a pfollowing frequencies:
FZERO = C1R1π 2
1
×××
FPOLE = R1Cπ 2
1
××× 2
FPOLE @origin = C1π 2 ××
1
Figures 7 and 8 show the gain and phase curves for the above transfer function with R1 = 9.3k, C1 = 1000pF, C2 = 100pF, and gm = .005 –1. It can be seen that at 50kHz, the error amplifier exhibits approximately 45° of
ar
Ω
phase m gin.
Figure 7. Error Amplifier Gain Curve
Figure 8. Error Amplifier Phase Curve
Total Open-Loop Response The open-loop response for the MIC2168A controller is easily obtained by adding the power path and the error amplifier gains together, since they already are in Log scale. It is desirable to have the gain curve intersect zero dB at tens of kilohertz, this is commonly called crossover frequency; the phase margin at crossover frequency should be at least 45°. Phase margins of 30° or less cause the power supply to have substantial ringing when subjected to transients, and have little tolerance for component or environmental variations. Figures 9 and 10 show th -loop gain and phase margin. It can be seen from Figure 9 that the gain curve intersects the 0dB at approximately 50kHz, and from Figure 10 that at 50kHz, the phase shows approximately 50° of margin.
e open
Micrel MIC2168A
January 2010 16 M9999-011510
Figure 9. Open-Loop Gain Margin
esistor ply to
lace a GND
the output capacitor ground. See demolayout, top layer.
10. The 10µF ceramic input cplaced between the drain osource of bottom MOSFET.
11. The 10µF ceramic capplaced right on the VDD pin without any vias.
12. The source of the bottom connect directly to the inwith a thick trace. The outhe input capacitor should
Figure 10. Open-Loop Phase Margin
Design Ex
ample Layout and Checklist:
1. Connect the current limiting (CS) rdirectly to the drain of top MOSFET Q1.
2. Use a 10Ω resistor from the input supthe VIN pin on the MIC2168A. Also, p1µF ceramic capacitor from this pin tonot through via.
3. The feedback resistors R1 and R2 should be placed close to the FB pin. The top side of R1 should connect directly to the output node. Run this trace away from the switch node (junction of Q1, Q2, and L1). The bottom side of R1 should connect to the GND pin on the
MIC2168A. 4. The compensation resistor and capacitors
should be placed right next to the COMP pin and the other side should connect directly to the GND pin on the MIC2168A rather than going to the plane.
5. Add a snubber circuit (resistor and a capacitor) from the switch node to GND. A good starting point is 1000pF and 1.4Ω.
6. Add a place holder for a gate resistor on the e drive. A gate resistors of
10 or less should be used. No gate resistor should be used on the low side MOSFET.
7. Low gate charge MOSFETs should be used to maximize efficiency, such as Si4800, Si4804BDY, IRF7821, IRF8910, FDS6680A, and FDS6912A to mention a few.
8. Add a 1Ω to 4Ω resistor from the SW pin on the MIC2168A to the switch node on the circuit (junction of MOSFETs and inductor).
9. Compensation component GND, feedback resistor ground, chip ground, 1µF VIN ceramic
and 10µF VDD capacitor should all run together and connect to
board
apacitor should be f top MOSFET and acitor should be
MOSFET should put capacitor GND tput capacitor and
connect directly to the GND plane.
13. Place a 0.1µF ceramic capacitor in parallel with the CS resistor to filter any switching noise.
top MOSFET gatΩ
capacitor ground,ground
Micrel MIC2168A
January 2010 17 M9999-011510
Recommended Land Pattern
Micrel MIC2168A
January 2010 18 M9999-011510
Package Information
10-Pin MSOP (MM)
MICREL, INC. 2180 FORTUNE DRIVE SAN JOSE, CA 95131 USA TEL +1 (408) 944-0800 FAX +1 (408) 474-1000 WEB http:/www.micrel.com
The information furnished by Micrel in this data sheet is believed to be accurate and reliable. However, no responsibility is assumed by Micrel for
its use. Micrel reserves the right to change circuitry and specifications at any time without notification to the customer.
Micrel Products are not designed or authorized for use as components in life support appliances, devices or systems where malfunction of a product can reasonably be expected to result in personal injury. Life support devices or systems are devices or systems that (a) are intended for
surgical implant into the body or (b) support or sustain life, and whose failure to perform can be reasonably expected to result in a significant injury to the user. A Purchaser’s use or sale of Micrel Products for use in life support appliances, devices or systems is a Purchaser’s own risk
and Purchaser agrees to fully indemnify Micrel for any damages resulting from such use or sale.
© 2003 Micrel, Incorporated.